26
© 2001 by CRC Press LLC 1 Introduction to MATLAB ® and Its Graphics Capabilities 1.1 Getting Started MATLAB can be thought of as a library of programs that will prove very use- ful in solving many electrical engineering computational problems. MAT- LAB is an ideal tool for numerically assisting you in obtaining answers, which is a major goal of engineering analysis and design. This program is very useful in circuit analysis, device design, signal processing, filter design, control system analysis, antenna design, microwave engineering, photonics engineering, computer engineering, and all other sub-fields of electrical engi- neering. It is also a powerful graphic and visualization tool. The first step in using MATLAB is to know how to call it. It is important to remember that although the front-end and the interfacing for machines with different operating systems are sometimes different, once you are inside MAT- LAB, all programs and routines are written in the same manner. Only those few commands that are for file management and for interfacing with external devices such as printers may be different for different operating systems. After entering MATLAB, you should see the prompt >> , which means the program interpreter is waiting for you to enter instructions. (Remember to press the Return key at the end of each line that you enter.) Now type clf . This command creates a graph window (if one does not already exist) or clears an existing graph window. Because it is impossible to explain the function of every MATLAB com- mand within this text, how would you get information on a certain command syntax? The MATLAB program has extensive help documentation available with simple commands. For example, if you wanted help on a function called roots (we will use this function often), you would type help roots . Note that the help facility cross-references other functions that may have related uses. This requires that you know the function name. If you want an idea of the available help files in MATLAB, type help . This gives you a list of topics included in MATLAB. To get help on a particular topic such as the Optimization Toolbox, type help toolbox/optim . This gives you a list of

Introduction to MATLAB® and Its Graphics Capabilitiessolen/Matlab/MatLab/Matlab...Introduction to MATLAB ® and Its Graphics Capabilities 1.1 Getting Started MATLAB can be thought

  • Upload
    others

  • View
    82

  • Download
    5

Embed Size (px)

Citation preview

1Introduction to MATLAB® andIts Graphics Capabilities

1.1 Getting Started

MATLAB can be thought of as a library of programs that will prove very use-ful in solving many electrical engineering computational problems. MAT-LAB is an ideal tool for numerically assisting you in obtaining answers,which is a major goal of engineering analysis and design. This program isvery useful in circuit analysis, device design, signal processing, filter design,control system analysis, antenna design, microwave engineering, photonicsengineering, computer engineering, and all other sub-fields of electrical engi-neering. It is also a powerful graphic and visualization tool.

The first step in using MATLAB is to know how to call it. It is important toremember that although the front-end and the interfacing for machines withdifferent operating systems are sometimes different, once you are inside MAT-LAB, all programs and routines are written in the same manner. Only thosefew commands that are for file management and for interfacing with externaldevices such as printers may be different for different operating systems.

After entering MATLAB, you should see the prompt >>, which means theprogram interpreter is waiting for you to enter instructions. (Remember topress the Return key at the end of each line that you enter.)

Now type clf. This command creates a graph window (if one does notalready exist) or clears an existing graph window.

Because it is impossible to explain the function of every MATLAB com-mand within this text, how would you get information on a certain commandsyntax? The MATLAB program has extensive help documentation availablewith simple commands. For example, if you wanted help on a function calledroots (we will use this function often), you would type help roots.

Note that the help facility cross-references other functions that may haverelated uses. This requires that you know the function name. If you want anidea of the available help files in MATLAB, type help. This gives you a listof topics included in MATLAB. To get help on a particular topic such as theOptimization Toolbox, type help toolbox/optim. This gives you a list of

0-8493-????-?/00/$0.00+$.50© 2000 by CRC Press LLC© 2001 by CRC Press LLC

all relevant functions pertaining to that area. Now you may type help forany function listed. For example, try help fmin.

1.2 Basic Algebraic Operations and Functions

The MATLAB environment can be used, on the most elementary level, as atool to perform simple algebraic manipulations and function evaluations.

Example 1.1Exploring the calculator functions of MATLAB. The purpose of this exampleis to show how to manually enter data and how to use basic MATLAB alge-braic operations. Note that the statements will be executed immediately afterthey are typed and entered (no equal sign is required).

Type and enter the text that follows the >> prompt to find out the MATLABresponses to the following:

2+2

5^2

2*sin(pi/4)

The last command gave the sine of π/4. Note that the argument of the functionwas enclosed in parentheses directly following the name of the function. There-fore, if you wanted to find sin3(π/4), the proper MATLAB syntax would be

sin(pi/4)^3

To facilitate its widespread use, MATLAB has all the standard elementarymathematical functions as built-in functions. Type help elfun, which isindexed in the main help menu to get a listing of some of these functions.Remember that this is just a small sampling of the available functions.

help elfun

The response to the last command will give you a large list of these elemen-tary functions, some of which may be new to you, but all of which will beused in your future engineering studies, and explored in later chapters ofthis book.

Example 1.2Assigning and calling values of parameters. In addition to inputting datadirectly to the screen, you can assign a symbolic constant or constants to rep-

© 2001 by CRC Press LLC

resent data and perform manipulations on them. For example, enter and notethe answer to each of the following:

a=2

b=3

c=a+b

d=a*b

e=a/b

f=a^3/b^2

g=a+3*b^2

Question: From the above, can you deduce the order in which MATLAB per-forms the basic operations?

In-Class Exercise

Pb. 1.1 Using the above values of a and b, find the values of:a. h = sin(a) sin(b)b. i = a1/3b3/7

c. j = sin–1(a/b) = arcsin(a/b)

1.3 Plotting Points

In this chapter section, you will learn how to use some simple MATLABgraphics commands to plot points. We use these graphics commands later inthe text for plotting functions and for visualizing their properties. To view allthe functions connected with 2-dimensional graphics, type:

help plot

All graphics functions connected with 3-dimensional graphics can be lookedup by typing

help plot3

A point P in the x-y plane is specified by two coordinates. The x-coordinatemeasures the horizontal distance of the point from the y-axis, while they-coordinate measures the vertical distance above the x-axis. These coordi-

© 2001 by CRC Press LLC

nates are called Cartesian coordinates, and any point in the plane can bedescribed in this manner. We write for the point, P(x, y).

Other representations can also be used to locate a point with respect to aparticular set of axes. For example, in the polar representation, the point isspecified by an r-coordinate that measures the distance of the point from theorigin, while the θ-coordinate measures the angle which the line passingthrough the origin and this point makes with the x-axis.

The purpose of the following two examples is to learn how to representpoints in a plane and to plot them using MATLAB.

Example 1.3Plot the point P(3, 4).

Solution: Enter the following:

x1=3;

y1=4;

plot(x1,y1,'*')

Note that the semicolon is used in the above commands to suppress theechoing of the values of the inputs. The '*' is used to mark the point that weare plotting. Other authorized symbols for point displays include 'o', '+','x', … the use of which is detailed in help plot.

Example 1.4Plot the second point, R(2.5, 4) on the graph while keeping point P of the pre-vious example on the graph.

Solution: If we went ahead, defined the coordinates of R, and attempted toplot the point R through the following commands:

x2=2.5;

y2=4;

plot(x2,y2,'o')

we would find that the last plot command erases the previous plot output.Thus, what should we do if we want both points plotted on the same

graph? The answer is to use the hold on command after the first plot.The following illustrates the steps that you should have taken instead of

the above:

hold on

x2=2.5;

© 2001 by CRC Press LLC

y2=4;

plot(x2,y2,'o')

hold off

The hold off turns off the hold on feature.

NOTES1. There is no limit to the number of plot commands you can type

before the hold is turned off.2. An alternative method for viewing multiple points on the same

graph is available: we may instead, following the entering of thevalues of x1, y1, x2, y2, enter:

plot(x1,y1,'*',x2,y2,'o')

This has the advantage, in MATLAB, of assigning automatically a differentcolor to each point.

1.3.1 Axes Commands

You may have noticed that MATLAB automatically adjusts the scale on agraph to accommodate the coordinates of the points being plotted. The axisscaling can be manually enforced by using the command axis([xminxmax ymin ymax]). Make sure that the minimum axis value is less than themaximum axis value or an error will result.

In addition to being able to adjust the scale of a graph, you can also changethe aspect ratio of the graphics window. This is useful when you wish to seethe correct x to y scaling. For example, without this command, a circle willlook more like an ellipse.

Example 1.5Plot the vertices of a square, keeping the geometric proportions unaltered.

Solution: Enter the following:

x1=-1;y1=-1;x2=1;y2=-1;x3=-1;y3=1;x4=1;y4=1;

plot(x1,y1,'o',x2,y2,'o',x3,y3,'o',x4,y4,'o')

axis([-2 2 -2 2])

axis square %square shape

Note that prior to the axis square command, the square looked like a rect-angle. If you want to go back to the default aspect ratio, type axis normal.The % symbol is used so that you can type comments in your program. Com-ments following the % symbol are ignored by the MATLAB interpreter.

© 2001 by CRC Press LLC

1.3.2 Labeling a Graph

To add labels to your graph, the functions xlabel, ylabel, and title canbe used as follows:

xlabel('x-axis')

ylabel('y-axis')

title('points in a plane')

If you desire to add a caption anywhere in the graph, you can use the MAT-LAB command gtext('caption') and place it at the location of yourchoice, on the graph, by clicking the mouse when the crosshair is properlycentered there.

1.3.3 Plotting a Point in 3-D

In addition to being able to plot points on a plane (2-D space), MATLAB isalso able to plot points in a three-dimensional space (3-D space). For this, weutilize the plot3 function.

Example 1.6Plot the point P(3, 4, 5).

Solution: Enter the following commands:

x1=3; y1=4; z1=5;

plot3(x1,y1,z1,'*')

You can also plot multiple points in a 3-D space in exactly the same way asyou did on a plane. Axis adjustment can still be used, but the vector inputinto the axis command must now have six entries, as follows:

axis([xmin xmax ymin ymax zmin zmax])

You can similarly label your 3-D figure using xlabel, ylabel, zlabel,and title.

1.4 M-files

In the last section, we found that to complete a figure with a caption, we hadto enter several commands one by one in the command window. Typing

© 2001 by CRC Press LLC

errors will be time-consuming to fix because if you are working in the com-mand window, you need to retype all or part of the program. Even if youdo not make any mistakes (!), all of your work may be lost if you inadvert-ently quit MATLAB and have not taken the necessary steps to save the con-tents of the important program that you just finished developing. Topreserve large sets of commands, you can store them in a special type of filecalled an M-file.

MATLAB supports two types of M-files: script and function M-files. To holda large collection of commands, we use a script M-file. The function M-file isdiscussed in Chapter 3. To make a script M-file, you need to open a file usingthe built-in MATLAB editor. For both Macs and PCs, first select New from thefile menu. Then select the M-file entry from the pull-down menu. After typingthe M-file contents, you need to save the file:

For Macs and PCs, select the save as command from the file win-dow. A field will pop up in which you can type in the name youhave chosen for this file (make sure that you do not name a file bya mathematical abbreviation, the name of a mathematical function,or a number). Also make sure that the file name has a .m extensionadded at the end of its name.

For Macs, save the file in a user’s designated volume.

For PCs, save the file in the default (bin) subdirectory.

To run your script M-file, just type the filename (omitting the .m extensionat its end) at the MATLAB prompt.

Example 1.7For practice, go to your file edit window to create the following file that youname myfile.m.

clear, clf

x1=1;y1=.5;x2=2;y2=1.5;x3=3;y3=2;

plot(x1,y1,'o',x2,y2,'+',x3,y3,'*')

axis([0 4 0 4])

xlabel('xaxis')

ylabel('yaxis')

title('3points in a plane')

After creating and saving myfile.m, go to the MATLAB command windowand enter myfile. MATLAB will execute the instructions in the order of thestatements stored in your myfile.m file.

© 2001 by CRC Press LLC

1.5 MATLAB Simple Programming

1.5.1 Iterative Loops

The power of computers lies in their ability to perform a large number ofrepetitive calculations. To do this without entering the value of a parameteror variable each time that these are changed, all computer languages havecontrol structures that allow commands to be performed and controlled bycounter variables, and MATLAB is no different. For example, the MATLAB“for” loop allows a statement or a group of statements to be repeated.

Example 1.8Generate the square of the first ten integers.

Solution: Edit and execute the the following script M-file:

for m=1:10

x(m)=m^2;

end;

In this case, the number of repetitions is controlled by the index variable m,which takes on the values m = 1 through m = 10 in intervals of 1. Therefore, tenassignments were made. What the above loop is doing is sequentially assign-ing the different values of m^2 (i.e., m2) in each element of the “x-array.” Anarray is just a data structure that can hold multiple entries. An array can be1-D such as in a vector, or 2-D such as in a matrix. More will be said aboutvectors and matrices in subsequent chapters. At this time, think of the 1-Dand 2-D arrays as pigeonholes with numbers or ordered pair of numbersrespectively assigned to them.

To find the value of a particular slot of the array, such as slot 3, enter:

x(3)

To read all the values stored in the array, type:

x

Question: What do you get if you enter m?

1.5.2 If-Else-End Structures

If a sequence of commands must be conditionally evaluated based on a rela-tional test, the programming of this logical relationship is executed withsome variation of an if-else-end structure.

© 2001 by CRC Press LLC

A. The simplest form of this structure is:if expression

commands evaluated if expression is True

else

commands evaluated if expression is False

end

NOTES1. The commands between the if and else statements are evaluated

if all elements in the expression are true.2. The conditional expression uses the Boolean logical symbols &

(and), | (or), and ~ (not) to connect different propositions.

Example 1.9Find for integer 0 < a ≤ 10, the values of C, defined as follows:

and b = 15.

Solution: Edit and execute the following script M-file:

for a=1:10

b=15;

if a>5

C(a)=a*b;

else

C(a)=(a*b)*(3/2);

end

end

Check that the values of C that you obtain by typing C are:

22.5 45 67.5 90 112.50 90 105 120 135 150

B. When there are three or more alternatives, the if-else-end struc-ture takes the form:if expression 1

Commands 1 evaluated if expression 1 is True

Cab a

ab a=

>

for

for

532

5

© 2001 by CRC Press LLC

elseif expression 2

Commands 2 evaluated if expression 2 is True

elseif expression 3

Commands 3 evaluated if expression 3 is True

else

Commands evaluated if no other expression is True

end

In this form, only the commands associated with the first True expressionencountered are evaluated; ensuing relational expressions are not tested.

1.5.2.1 Alternative Syntax to the if Statement

As an alternative to the if syntax, we can use, in certain instances, Booleanexpressions to specify an expression in different domains. For example,(x>=l) has the value 1 if x is larger than or equal to 1 and zero otherwise;and (x<=h) is equal to 1 when x is smaller than or equal to h, and zero oth-erwise.

The relational operations allowed inside the parentheses are: ==, <=, >=,~=, <, >.

Homework Problem

Pb. 1.2 For the values of integer a going from 1 to 10, using separately themethods of the if syntax and the Boolean alternative expressions, find thevalues of C if:

Use the stem command to graphically show C.

1.6 Array Operations

In the above examples, we used for loops repeatedly. However, this kind ofloop-programming is very inefficient and must be avoided as much as possi-

C a a

C a a

C a a

= <= + ≤ <= ≥

2 35 3 7

7

forforfor

© 2001 by CRC Press LLC

ble in MATLAB. In fact, ideally, a good MATLAB program will always mini-mize the use of loops because MATLAB is an interpreted language — not acompiled one. As a result, any looping process is very inefficient. Neverthe-less, at times we use the for loops, when necessitated by pedagogical reasons.

To understand array operations more clearly, consider the following:

a=1:3 % a starts at 1, goes to 3 in increments of 1.

If the increment is not 1, you must specify the increment; for example:

b=2:2:6 % b starts at 2, goes to 6 in increments of 2

To distinguish arrays operations from either operations on scalars or onmatrices, the symbol for multiplication becomes .*, that of division ./, andthat of exponentiation .^. Thus, for example:

c=a.*b % takes every element of a and multiplies% it by the element of b in the same array location

Similarly, for exponentiation and division:

d=a.^b

e=a./b

If you try to use the regular scalar operations symbols, you will get an errormessage.

Note that array operations such as the above require that the two arrayshave the same length (i.e., the same number of elements). To verify that twoarrays have the same number of elements (dimension), use the length com-mand. Thus, to find the length of a and b, enter:

length(a)

length(b)

NOTE The expression x=linspace(0,10,200) is also the generator foran x-array with first element equal to 0, a last element equal to 10, and having200 equally spaced points between 0 and 100. Here, the number of pointsrather than the increment is specified; that is, length(x)=200.

1.7 Curve and Surface Plotting

Review the sections of the Supplement pertaining to lines, quadratic func-tions, and trigonometric functions before proceeding further.

© 2001 by CRC Press LLC

1.7.1 x-y Parametric Plot

Now edit another M-file called myline.m as follows and execute it.

N=10;

for m=1:N

x(m)=m;

y(m)=2*m+3;

end

plot(x,y)

After executing the M-file using myline, you should see a straight line con-necting the points (1, 5) and (10, 23). This demonstration shows the basic con-struct for creating two arrays and plotting the points with their x-coordinatefrom a particular location in one array and their y-coordinate from the samelocation in the second array. We say that the plot command here plotted they-array vs. the x-array.

We note that the points are connected by a continuous line making asmooth curve; we say that the program graphically interpolated the discretepoints into a continuous curve. If we desire to see additionally the individualpoints corresponding to the values of the arrays, the last command should bechanged to:

plot(x,y,x,y,'o')

Example 1.10Plot the two curves y1 = 2x + 3 and y2 = 4x + 3 on the same graph.

Solution: Edit and execute the following script M-file:

for m=1:10 m=1:10;

x(m)=m; x=m;

y1(m)=2*m+3; or better y1=2*m+3;

y2(m)=4*m+3; y2=4*m+3;

end plot(x,y1,x,y2)

plot(x,y1,x,y2)

Finally, note that you can separate graphs in one figure window. This isdone using the subplot function in MATLAB. The arguments of the subplotfunction are subplot(m,n,p), where m is the number of rows partitioningthe graph, n is the number of columns, and p is the particular subgraph cho-sen (enumerated through the left to right, top to bottom convention).

© 2001 by CRC Press LLC

1.7.1.1 Demonstration: Plotting Multiple Figures within a Figure Window

Using the data obtained in the previous example, observe the difference inthe partition of the page in the following two sets of commands:

subplot(2,1,1)

plot(x,y1)

subplot(2,1,2)

plot(x,y2)

and

clf

subplot(1,2,1)

plot(x,y1)

subplot(1,2,2)

plot(x,y2)

1.7.2 More on Parametric Plots in 2-D

In the preceding subsection, we generated the x- and y-arrays by first writingthe x-variable as a linear function of a parameter, and then expressed thedependent variable y as a function of that same parameter. What we did isthat, instead of thinking of a function as a relation between an independentvariable x and a dependent variable y, we thought of both x and y as beingdependent functions of a third independent parameter. This method of curverepresentation, known as the parametric representation, is described by (x(t),y(t)), where the parameter t varies over some finite domain (tmin, tmax). Note,however, that in the general case, unlike the examples in the previous chaptersubsection, the independent variable x need not be linear in the parameter,nor is the process of parametrization unique.

Example 1.11Plot the trigonometric circle.

Solution: Recalling that the x-coordinate of any point on the trigonometriccircle has the cosine as x-component and the sine as y-component, the gener-ation of the trigonometric circle is immediate:

th=linspace(0,2*pi,101)

x=cos(th);

y=sin(th);

© 2001 by CRC Press LLC

plot(x,y)

axis square

The parametric representation of many common curves is our next topic ofinterest. The parametric representation is defined such that if x and y are con-tinuous functions of t over the interval I, we can describe a curve in the x-yplane by specifying:

C: x = x(t), y = y(t), and t ∈ I

More Examples:In the following examples, we want to identify the curves f(x, y) = 0 corre-sponding to each of the given parametrizations.

Example 1.12C: x = 2t – 1, y = t + 1, and 0 < t < 2. The initial point is at x = –1, y = 1, and thefinal point is at x = 3, y = 3.

Solution: The curve f(x, y) = 0 form can be obtained by noting that:

2t – 1 = x ⇒ t = (x + 1)/2

Substitution into the expression for y results in:

This describes a line with slope 1/2 crossing the x-axis at x = –3.

Question: Where does this line cross the y-axis?

Example 1.13C: x = 3 + 3 cos(t), y = 2 + 2 sin(t), and 0 < t < 2π. The initial point is at x = 6, y= 2, and the final point is at x = 6, y = 2.

Solution: The curve f(x, y) = 0 can be obtained by noting that:

Using the trigonometric identity cos2(t) + sin2(t) = 1, we deduce the followingequation:

yx= +2

32

sin( ) costy

tx=

− ( ) = −22

33

and

© 2001 by CRC Press LLC

This is the equation of an ellipse centered at x = 3, y = 2 and having major andminor radii equal to 3 and 2, respectively.

Question 1: What are the coordinates of the foci of this ellipse?

Question 2: Compare the above curve with the curve defined through:

x = 3 + 3 cos(2t), y = 2 + 2 sin(2t), and 0 < t < 2π

What conclusions can you draw from your answer?

In-Class Exercises

Pb. 1.3 Show that the following parametric equations:

x = h + a sec(t), y = k + b tan(t), and –π/2 < t < π/2

are those of the hyperbola also represented by the equation:

Pb. 1.4 Plot the hyperbola represented by the parametric equations of Pb.1.3, with h = 2, k = 2, a = 1, b = 2. Find the coordinates of the vertices and thefoci. (Hint: One branch of the hyperbola is traced for –π/2 < t < π/2, while theother branch is traced when π/2 < t < 3π/2.)

Pb. 1.5 The parametric equations of the cycloid are given by:

x = Rωt + R sin(ωt), y = R + R cos(ωt), and 0 < t

Show how this parametric equation can be obtained by following the kine-matics of a point attached to the outer rim of a wheel that is uniformly rolling,without slippage, on a flat surface. Relate the above parameters to the linearspeed and the radius of the wheel.

Pb. 1.6 Sketch the curve C defined through the following parametric equa-tions:

( ) ( )y x−+ − =

22

33

12

2

2

2

( ) ( )x ha

y kb

− −−

=2

2

2

2 1

© 2001 by CRC Press LLC

Homework Problems

The following set of problems provides the mathematical basis for under-standing the graphical display on the screen of an oscilloscope, when in thex-y mode.

Pb. 1.7 To put the quadratic expression

Ax2 + Bxy + Cy2 + Dx + Ey + F = 0

in standard form (i.e., to eliminate the x-y mixed term), make the transformation

Show that the mixed term is eliminated if

Pb. 1.8 Consider the parametric equations

C: x = a cos(t), y = b sin(t + ϕ), and 0 < t < 2π

where the initial point is at x = a, y = b sin(ϕ), and the final point is at x = a,y = b sin(ϕ).

a. Obtain the equation of the curve in the form f(x, y) = 0.b. Using the results of Pb. 1.7, prove that the ellipse inclination angle

is given by:

x t

t t

t t

t t

y t

t

t t

( )

tan ( )

tan ( )

( )

tan ( )

tan

=+ − ≤ ≤ −

+ − −

− < <

− + −

< <

=− ≤ ≤ −

− < <

2 3 1

113 3

1 1 0

113 3

1 0 1

0 3 113 3

1 1 0

13

2

2

2

for

for

for

for

for

π

π

π

ππ3

1 0 12( ) −

< <

t tfor

x x y

y x y

= ′ − ′

= ′ + ′

cos( ) sin( )

sin( ) cos( )

θ θ

θ θ

cot( )( )

.2θ ≡ −A CB

cot( )( )

sin( )2

2

2 2

θϕ

≡ −a bab

© 2001 by CRC Press LLC

Pb. 1.9 If the parametric equations of a curve are given by:

C: x = cos(t), y = sin(2t), and 0 < t < 2π

where the initial point is at x = 1, y = 0, and the final point is at x = 1, y = 0.The curve so obtained is called a Lissajous figure. It has the shape of a fig-

ure 8 with two nodes in the x-direction and only one node in the y-direction.What do you think the parametric equations should be if we wanted m

nodes on the x-axis and n nodes on the y-axis? Test your hypothesis by plot-ting the results.

1.7.3 Plotting a 3-D Curve

Our next area of exploration is plotting 3-D curves.

Example 1.14Plot the helix.

Solution: To plot a helical curve, we can imagine initially that a point isrevolving at a uniform speed around the perimeter of a circle. Now imaginethat as the circular motion is continuing, the point is moving away from thex-y plane at some constant linear speed. The parametric representation of thismotion can be implemented in MATLAB through the following:

for m=1:201

th(m)=2*pi*.01*(m-1);

x(m)=cos(th(m));

y(m)=sin(th(m));

z(m)=th(m);

end

plot3(x,y,z)

In-Class Exercises

Pb. 1.10 In the helix of Example 1.14, what is the vertical distance (the pitch)between two consecutive helical turns. How can you control this distance?Find two methods of implementation.

Pb. 1.11 If instead of a circle in 2-D, as in the helix, the particle describes in2-D a Lissajous pattern having two nodes in the y-direction and three nodes

© 2001 by CRC Press LLC

in the x-direction, assuming that the z-parametric equation remains the same,show the resulting 3-D trajectory.

Pb. 1.12 What if z(t) is periodic in t? For example, z(t) = cos(t) or z(t) =cos(2t), while the 2-D motion is still circular. Show the 3-D trajectory.

In Example 1.14, we used the for loop to generate the dependent arrays forthe helix; but as pointed out previously, a more efficient method to programthe helix is in the array notation, as follows:

th=[0:.01:2]*2*pi;

x=cos(th);

y=sin(th);

z=th;

plot3(x,y,z)

1.7.4 Plotting a 3-D Surface

We now explore the two different techniques for rendering, in MATLAB, 3-Dsurface graphics: the mesh and the contour representations.

• A function of two variables z = f(x, y) represents a surface in 3-Dgeometry; for example:

z = ax + by + c

represents a plane that crosses the vertical axis (z-axis) at c.• There are essentially two main techniques in MATLAB for viewing

surfaces: the mesh function and the contour function.• In both techniques, we must first create a 2-D array structure (like

a checkerboard) with the appropriate x- and y-values. To imple-ment this, we use the MATLAB meshgrid function.

• The z-component is then expressed in the variables assigned toimplement the meshgrid command.

• We then plot the function with either the mesh command or thecontour command. The mesh command gives a 3-D rendering ofthe surface, while the contour command gives contour lines,wherein each contour represents the locus of points on the surfacehaving the same height above the x-y plane. This last renderingtechnique is that used by mapmakers to represent the topographyof a terrain.

© 2001 by CRC Press LLC

1.7.4.1 Surface Rendering

Example 1.15Plot the sinc function whose equation is given by:

over the domain –8 < x < 8 and –8 < y < 8.

Solution: The implementation of the mesh rendering follows:

x=[-8:.1:8];

y=[-8:.1:8];

[X,Y]=meshgrid(x,y);

R=sqrt(X.^2+Y.^2)+eps;

Z=sin(R)./R;

mesh(X,Y,Z)

The variable eps is a tolerance number = 2–52 used for determining expres-sions near apparent singularities, to avoid numerical division by zero.

To generate a contour plot, we replace the last command in the above by:

contour(X,Y,Z,50) % The fourth argument specifies% the number of contour lines to be shown

If we are interested only in a particular contour level, for example, the onewith elevation Z0, we use the contour function with an option, as follows:

contour(X,Y,Z,[Zo Zo])

Occasionally, we might be interested in displaying simultaneously themesh and contour rendering of a surface. This is possible through the use ofthe command meshc. It is the same as the mesh command except that acontour plot is drawn beneath the mesh.

Preparatory Activity: Look in your calculus book for some surfaces equations,such as those of the hyperbolic paraboloid and the elliptic paraboloid andothers of your choice for the purpose of completing Pb. 1.16 of the next in-class activity.

zx y

x y=

+( )+

sin 2 2

2 2

© 2001 by CRC Press LLC

In-Class Exercises

Pb. 1.13 Use the contour function to graphically find the locus of pointson the above sinc surface that are 1/2 units above the x-y plane (i.e., the sur-face intersection with the z = 1/2 plane).

Pb. 1.14 Find the x-y plane intersection with the following two surfaces:

Pb. 1.15 Verify your answers to Pb. 1.14 with that which you would obtainanalytically for the shape of the intersection curves of the surfaces with thex-y plane. Also, compute the coordinates of the point of intersection of thetwo obtained curves. Verify your results graphically.

Pb. 1.16 Plot the surfaces that you have selected in your preparatory activ-ity. Look in the help folder for the view command to learn how to view thesesurfaces from different angles.

1.8 Polar Plots

MATLAB can also display polar plots. In the first example, we draw anellipse of the form r = 1 + ε cos(θ) in a polar plot; other shapes are given in theother examples.

Example 1.16Plot the ellipse in a polar plot.

Solution: The following sequence of commands plot the polar plot of anellipse with ε = 0.2:

th=0:2*pi/100:2*pi;

rho=1+.2*cos(th);

polar(th,rho)

The shape you obtain may be unfamiliar; but to verify that this is indeed anellipse, view the curve in a Cartesian graph. For that, you can use the MAT-LAB polar to Cartesian converter pol2cart, as follows:

z x y

z x y

1

2

3

4 2 4

= + +

= − −

© 2001 by CRC Press LLC

[x,y]=pol2cart(th,rho);

plot(x,y)

axis equal

Example 1.17Graph the polar plot of a spiral.

Solution: The equation of the spiral is given by:

r = aθ

Its polar plot can be viewed by executing the following script M-file (a = 3):

th=0:2*pi/100:2*pi;

rho=3*th;

polar(th,rho)

In-Class Exercises

Pb. 1.17 Prove that the polar equation r = 1 + ε cos(θ), where ε is alwaysbetween –1 and 1, results in an ellipse. (Hint: Relate ε to the ratio between thesemi-major and semi-minor axis.) It is worth noting that the planetary orbitsare usually described in this manner in most astronomy books.

Pb. 1.18 Plot the three curves described by the following polar equations:

Pb. 1.19 Plot:

r = sin(2θ) cos(2θ)

The above gives a flower-type curve with eight petals. How would you makea flower with 16 petals?

Pb. 1.20 Plot:

r = sin2(θ)

This two-lobed structure shows the power distribution of a simple dipoleantenna. Note the directed nature of the radiation. Can you increase thedirectivity further?

r r r= − = − =2 2 1 2 2 2sin( ), sin( ), sin( )θ θ θ

© 2001 by CRC Press LLC

Pb. 1.21 Acquaint yourself with the polar plots of the following curves:(choose first a = 1, then experiment with other values).

a. Straight lines:

b. Cissoid of Diocles:

c. Strophoid:

d. Folium of Descartes:

1.9 Animation

A very powerful feature of MATLAB is its ability to render an animation. Forexample, suppose that we want to visualize the oscillations of an ordinaryspring. What are the necessary steps to implement this objective?

1. Determine the parametric equations that describe the curve at afixed time. In this instance, it is the helix parametric equations asgiven earlier in Example 1.14.

2. Introduce the time dependence in the appropriate curve parame-ters. In this instance, make the helix pitch to be oscillatory in time.

3. Generate 3-D plots of the curve at different times. Make sure thatyour axis definition includes all cases.

4. Use the movie commands to display consecutively the differentframes obtained in step 3.

The following script M-file implements the above workplan:

th=0:pi/60:32*pi;

a=1;

A=0.25;

w=2*pi/15;

M=moviein(16);

for t=1:16;

x=a*cos(th);

ra

=+

≤ ≤10

2cos( ) sin( )

θ θθ π

for

r a= − ≤ ≤sin ( )cos( )

2

3 3θθ

π θ πfor

ra= − ≤ ≤cos( )

cos( )

23 3

θθ

π θ πfor

ra=

+− ≤ ≤3

6 23 3

sin( )cos( )sin ( ) cos ( )

θ θ

θ θπ θ π

for

© 2001 by CRC Press LLC

y=a*sin(th);

z=(1+A*cos(w*(t-1)))*th;

plot3(x,y,z,'r');

axis([-2 2 -2 2 0 40*pi]);

M(:,t)=getframe;

end

movie(M,15)

The statement M=moviein(16) creates the 2-D structure that stores ineach column the data corresponding to a frame at a specific time. The framesthemselves are generated within the for loop. The getframe functionreturns a pixel image of the image of the different frames. The last commandplays the movie n-times (15, in this instance).

1.10 Histograms

The most convenient representation for data collected from experiments is inthe form of histograms. Typically, you collect data and want to sort it out indifferent bins; the MATLAB command for this operation is hist. But prior togetting to this point, let us introduce some array-related definitions and learnthe use of the MATLAB commands that compute them.

Let {yn} be a data set; it can be represented in MATLAB by an array. Thelargest element of this array is obtained through the command max(y), andthe smallest element is obtained through the command min(y).

The mean value of the elements of the array is obtained through the com-mand mean(y), and the standard deviation is obtained through the com-mand std(y).

The definitions of the mean and of the standard deviation are, respectively,given by:

where N is the dimension of the array.

y

y i

N

N y i y i

N N

i

N

yi

N

i

N

=

=

=

= =

∑ ∑

( )

( ( )) ( )

( )

1

2

1 1

2

© 2001 by CRC Press LLC

The data (i.e., the array) can be organized into a number of bins (nb) andexhibited through the command [n,y]=hist(y,nb); the array n in theoutput will be the number of elements in each of the bins.

Example 1.18Find the mean and the standard deviation and draw the histogram, with 20bins, for an array whose 10,000 elements are chosen from the MATLAB built-in normal distribution with zero mean and standard deviation 1.

Solution: Edit and execute the following script M-file:

y=randn(1,10000);

meany=mean(y)

stdy=std(y)

nb=20;

hist(y,nb)

You will notice that the results obtained for the mean and the standard devi-ation vary slightly from the theoretical results. This is due to the finite num-ber of elements chosen for the array and the intrinsic limit in the built-inalgorithm used for generating random numbers.

NOTE The MATLAB command for generating an N-elements array of ran-dom numbers generated uniformly from the interval [0, 1] is rand(1,N).

1.11 Printing and Saving Work in MATLAB

Printing a figure: Use the MATLAB print function to print a displayed figuredirectly to your printer. Notice that the printed figure does not take up theentire page. This is because the default orientation of the graph is in portraitmode. To change these settings, try the following commands on an alreadygenerated graphic window:

orient('landscape') %full horizontal layout

orient('tall') %full vertical layout

Printing a program file (script M-file): For both the Mac and PC, open the M-filethat you want to print. Go to the File pull-down menu, and select Print.

Saving and loading variables (data): You can use the MATLAB save function toeither save a particular variable or the entire MATLAB workspace. To do this,follow the following example:

© 2001 by CRC Press LLC

x=1;y=2;

save 'user volume:x'

save 'user volume:workspace'

The first save command saved the variable x into a file x.mat. You canchange the name of the .mat file so it does not match the variable name, butthat would be confusing. The second command saves all variables (x and y)in the workspace into workspace.mat.

To load x.mat and workspace.mat, enter MATLAB and use the MATLABload functions; note what you obtain if you entered the following commands:

load 'user volume:x'

x

load 'user volume:workspace'

y

After loading the variables, you can see a list of all the variables in yourworkplace if you enter the MATLAB who command.

What would you obtain if you had typed and entered the who command atthis point?

Now, to clear the workspace of some or all variables, use the MATLABclear function.

clear x %clears variable x from the workspace

clear %clears all variables from workspace

1.12 MATLAB Commands Review

axis Sets the axis limits for both 2-D and 3-D plots. Axissupports the arguments equal and square, whichmakes the current graphs aspect ratio 1.

contour Plots contour lines of a surface.clear Clears all variables from the workspace.clf Clears figure.for Runs a sequence of commands a given number of

times.getframe Returns the pixel image of a movie frame.help Online help.hold on(off) Holds the plot axis with existing graphics on, so that

multiple figures can be plotted on the same graph(release the hold of the axes).

© 2001 by CRC Press LLC

if Conditional evaluation.length Gives the length of an array.load Loads data or variable values from previous sessions

into current MATLAB session.linspace Generates an array with a specified number of points

between two values.meshgrid Makes a 2-D array of coordinate squares suitable for

plotting surface meshes.mesh Plots a mesh surface of a surface stored in a matrix.meshc The same as mesh, but also plots in the same figure

the contour plot.min Finds the smallest element of an array.max Finds the largest element of an array.mean Finds the mean of the elements of an array.moviein Creates the matrix that contains the frames of an ani-

mation.movie Plays the movie described by a matrix M.orient Orients the current graph to your needs.plot Plots points or pairs of arrays on a 2-D graph.plot3 Plots points or array triples on a 3-D graph.polar Plots a polar plot on a polar grid.pol2cart Polar to Cartesian conversion.print Prints a figure to the default printer.quit or exit Leave MATLAB program.rand Generates an array with elements randomly chosen

from the uniform distribution over the interval [0, 1].randn Generates an array with elements randomly chosen

from the normal distribution function with zeromean and standard deviation 1.

subplot Partitions the graphics window into sub-windows.save Saves MATLAB variables.std Finds the standard deviation of the elements of an

array.stem Plots the data sequence as stems from the x-axis ter-

minated with circles for the data value.view Views 3-D graphics from different perspectives.who Lists all variables in the workspace.xlabel, ylabel,zlabel, title

Labels the appropriate axes with text and title.

(x>=x1) Boolean function that is equal to 1 when the conditioninside the parenthesis is satisfied, and zero otherwise.

© 2001 by CRC Press LLC